[lldb] Use validation combination of options in TestAbbreviations

[JDevlieghere]

Name and line number are part of different option groups and are not compatible.

(lldb) breakpoint set -n foo -l 10
error: invalid combination of options for the given command

The help output for breakpoint set confirms this. This patch updates the test to use two compatible options. With the improved error reporting from #125125 this becomes an issue.

[llvmbot]

@llvm/pr-subscribers-lldb

Author: Jonas Devlieghere (JDevlieghere)

Changes

Name and line number are part of different option groups and are not compatible.

(lldb) breakpoint set -n foo -l 10
error: invalid combination of options for the given command

The help output for breakpoint set confirms this. This patch updates the test to use two compatible options. With the improved error reporting from #125125 this becomes an issue.

---

Full diff: https://github.com/llvm/llvm-project/pull/125270.diff

1 Files Affected:

• (modified) lldb/test/API/functionalities/abbreviation/TestAbbreviations.py (+4-4)

diff --git a/lldb/test/API/functionalities/abbreviation/TestAbbreviations.py b/lldb/test/API/functionalities/abbreviation/TestAbbreviations.py
index a8cbffbb7ba4a5..bef159501c6cd1 100644
--- a/lldb/test/API/functionalities/abbreviation/TestAbbreviations.py
+++ b/lldb/test/API/functionalities/abbreviation/TestAbbreviations.py
@@ -45,10 +45,10 @@ def test_command_abbreviations_and_aliases(self):
 
         # Make sure an unabbreviated command is not mangled.
         command_interpreter.ResolveCommand(
-            "breakpoint set --name main --line 123", result
+            "breakpoint set --name main --ignore-count 123", result
         )
         self.assertTrue(result.Succeeded())
-        self.assertEqual("breakpoint set --name main --line 123", result.GetOutput())
+        self.assertEqual("breakpoint set --name main --ignore-count 123", result.GetOutput())
 
         # Create some aliases.
         self.runCmd("com a alias com al")
@@ -72,10 +72,10 @@ def test_command_abbreviations_and_aliases(self):
             "process launch -s -o /dev/tty0 -e /dev/tty0", result.GetOutput()
         )
 
-        self.runCmd("alias xyzzy breakpoint set -n %1 -l %2")
+        self.runCmd("alias xyzzy breakpoint set -n %1 -i %2")
         command_interpreter.ResolveCommand("xyzzy main 123", result)
         self.assertTrue(result.Succeeded())
-        self.assertEqual("breakpoint set -n main -l 123", result.GetOutput().strip())
+        self.assertEqual("breakpoint set -n main -i 123", result.GetOutput().strip())
 
         # And again, without enough parameters.
         command_interpreter.ResolveCommand("xyzzy main", result)

[github-actions]

✅ With the latest revision this PR passed the Python code formatter.

[jimingham]

I'm happy that you can make ResolveCommand fail on commands that aren't valid! These tests clearly thought they were making up a legal command string, so this change expresses better what the test intends anyway.